[GitHub] szha commented on a change in pull request #7494: gluon save/load optimizer states

[git]

szha commented on a change in pull request #7494: gluon save/load optimizer 
states
URL: https://github.com/apache/incubator-mxnet/pull/7494#discussion_r136909694
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -969,12 +969,21 @@ def sync_state_context(self, state, context):
 
 def set_states(self, states):
 """Sets updater states."""
-self.states = pickle.loads(states)
+states = pickle.loads(states)
+if isinstance(states, tuple) and len(states) == 2:
 
 Review comment:
   Thanks for the comment. What's your suggestion @ptrendx?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ptrendx commented on a change in pull request #7494: gluon save/load optimizer states

[git]

ptrendx commented on a change in pull request #7494: gluon save/load optimizer 
states
URL: https://github.com/apache/incubator-mxnet/pull/7494#discussion_r136910105
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -969,12 +969,21 @@ def sync_state_context(self, state, context):
 
 def set_states(self, states):
 """Sets updater states."""
-self.states = pickle.loads(states)
+states = pickle.loads(states)
+if isinstance(states, tuple) and len(states) == 2:
 
 Review comment:
   Not sure (especially since we need to keep backwards compatibility). Use 
dictionary instead of tuple for dumping both state and optimizer?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7494: gluon save/load optimizer states

[git]

szha commented on a change in pull request #7494: gluon save/load optimizer 
states
URL: https://github.com/apache/incubator-mxnet/pull/7494#discussion_r136912002
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -969,12 +969,21 @@ def sync_state_context(self, state, context):
 
 def set_states(self, states):
 """Sets updater states."""
-self.states = pickle.loads(states)
+states = pickle.loads(states)
+if isinstance(states, tuple) and len(states) == 2:
 
 Review comment:
   That sounds more readable, though fundamentally the agreement on keys in 
dict isn't any stronger, more extensible, or more robust than type constraints 
and ordering of elements.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7494: gluon save/load optimizer states

[git]

szha commented on a change in pull request #7494: gluon save/load optimizer 
states
URL: https://github.com/apache/incubator-mxnet/pull/7494#discussion_r136912489
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -969,12 +969,21 @@ def sync_state_context(self, state, context):
 
 def set_states(self, states):
 """Sets updater states."""
-self.states = pickle.loads(states)
+states = pickle.loads(states)
+if isinstance(states, tuple) and len(states) == 2:
 
 Review comment:
   That said, I'm open to a change with more thoughts to it.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ptrendx commented on a change in pull request #7494: gluon save/load optimizer states

[git]

ptrendx commented on a change in pull request #7494: gluon save/load optimizer 
states
URL: https://github.com/apache/incubator-mxnet/pull/7494#discussion_r136913157
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -969,12 +969,21 @@ def sync_state_context(self, state, context):
 
 def set_states(self, states):
 """Sets updater states."""
-self.states = pickle.loads(states)
+states = pickle.loads(states)
+if isinstance(states, tuple) and len(states) == 2:
 
 Review comment:
   Optimizers already use tuples for states, so in that regard any type is 
better than tuple here. I would also argue that since fields in dictionary are 
named with human readable text it makes it much less error prone.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha opened a new pull request #7732: Update optimizer.py

[git]

szha opened a new pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7494: gluon save/load optimizer states

[git]

szha commented on a change in pull request #7494: gluon save/load optimizer 
states
URL: https://github.com/apache/incubator-mxnet/pull/7494#discussion_r136916902
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -969,12 +969,21 @@ def sync_state_context(self, state, context):
 
 def set_states(self, states):
 """Sets updater states."""
-self.states = pickle.loads(states)
+states = pickle.loads(states)
+if isinstance(states, tuple) and len(states) == 2:
 
 Review comment:
   Agreed. Proposed change in #7732
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ChidanandKumarKS opened a new issue #7733: How to convert the data of the result of mx.symbol.Convolution to nd.array data format?

[git]

ChidanandKumarKS opened a new issue #7733: How to convert the data of the 
result of mx.symbol.Convolution to nd.array data format?
URL: https://github.com/apache/incubator-mxnet/issues/7733
 
 
   How to convert the data of the result of mx.symbol.Convolution to nd.array 
data format?
   Kindly suggest
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] NiklasDL opened a new issue #7734: Error in mx.init.internal.default(name, shape, ctx, allow.unknown) : Unkown initialization pattern for identityattachklsparsereg0_moving_avg

[git]

NiklasDL opened a new issue #7734: Error in mx.init.internal.default(name, 
shape, ctx, allow.unknown) :  Unkown initialization pattern for  
identityattachklsparsereg0_moving_avg
URL: https://github.com/apache/incubator-mxnet/issues/7734
 
 
   Operating System: Windows 
   
   Compiler:
   
   Package used: Rversion:
   
   Or if installed from source:
   
   MXNet commit hash (`git rev-parse HEAD`):
   
   If you are using python package, please provide
   
   Python version and distribution:
   
   If you are using R package, please provide
   
   R `sessionInfo()`:
   
   ## Error Message:
   Please paste the full error message, including stack trace.
   
   ## Minimum reproducible example
   if you are using your own code, please provide a short script that 
reproduces the error.
   
   ## Steps to reproduce
   or if you are running standard examples, please provide the commands you 
have run that lead to the error.
   
   1.
   2.
   3.
   
   ## What have you tried to solve it?
   
   1.
   2.
   3.
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] chowkamlee81 opened a new issue #7735: My jupyter notebook uses old mxnet version 0.9.5 whereas pycharm debugger uses 0.11.1...

[git]

chowkamlee81 opened a new issue #7735: My jupyter notebook uses old mxnet 
version 0.9.5 whereas pycharm debugger uses 0.11.1...
URL: https://github.com/apache/incubator-mxnet/issues/7735
 
 
   Recently i deleted mxnet version 0.9.5 and upgraded to 0.11.1 version
   Still my jupyter notebook points to old version0.9.5 rather than new one 
0.11.1.
   Kindly help
   
   Chowkamlee
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ptrendx commented on issue #7708: mx.io.ImageRecordIter confused me, does this function is related to time?

[git]

ptrendx commented on issue #7708: mx.io.ImageRecordIter confused me, does this 
function is related to time?
URL: 
https://github.com/apache/incubator-mxnet/issues/7708#issuecomment-327121142
 
 
   Ok, I did reproduce it. Will debug it now.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] chowkamlee81 opened a new issue #7736: Formula to calculate dimensions of output of Deconvolution filter given kernel filter size and stride.

[git]

chowkamlee81 opened a new issue #7736: Formula to calculate dimensions of 
output of Deconvolution filter given kernel filter size and stride.
URL: https://github.com/apache/incubator-mxnet/issues/7736
 
 
   Formula to calculate dimensions of output of Deconvolution filter given 
kernel filter size and stride 
   How the stride factor is related to the upsampling factor?
   Kindly suggest
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] Rizhiy commented on issue #7631: windows MXNET_CUDNN_AUTOTUNE_DEFAULT = 0 no work.

[git]

Rizhiy commented on issue #7631: windows MXNET_CUDNN_AUTOTUNE_DEFAULT = 0 no 
work.
URL: 
https://github.com/apache/incubator-mxnet/issues/7631#issuecomment-327127112
 
 
   Happening for me as well on Ubuntu.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji opened a new pull request #7737: Use memcpy instead of assigning each individual element

[git]

yaochengji opened a new pull request #7737: Use memcpy instead of assigning 
each individual element
URL: https://github.com/apache/incubator-mxnet/pull/7737
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ptrendx commented on issue #7708: mx.io.ImageRecordIter confused me, does this function is related to time?

[git]

ptrendx commented on issue #7708: mx.io.ImageRecordIter confused me, does this 
function is related to time?
URL: 
https://github.com/apache/incubator-mxnet/issues/7708#issuecomment-327134873
 
 
   Ok, I understand why it happens:
   reset() function in Python calls BeforeFirst function in C interface, which 
is supposed to reset the iterator. BeforeFirst function is also used after the 
epoch ends to reset the iterator. Since the dataset size may be indivisible by 
the batch size (and it is in the example shown by @chenboheng), by default 
ImageRecordIter resets itself and takes images from the next epoch to fill the 
last batch (it is shown in the first output shown in this issue - 2 last images 
from the last batch are the same as 2 first images of the first batch). Then 
the "reset" of iterator does not actually reset it - if it was, it would read 
those 2 images twice. 
   In order to improve performance, ImageRecordIter prefetches a few batches of 
data. It turns out that this dataset is small enough that the iterator 
prefetches all of it in those 2 seconds when the main thread sleeps, and so it 
reaches the end of epoch internally. Calling reset in that situation calls 
BeforeFirst in the same way it would be called after epoch is done, so it does 
not actually reset the iterator, but prepares it for a new epoch, with dataset 
being offset by those 2 images that it used to fill the last batch of the 
"previous epoch". That is why the result is different.  
   So basically the bug can only be observed if the dataset is small enough 
that it fits in internal prefetch queue, the size of dataset is not divisible 
by the batch size and we give it enough time to internally process the entire 
dataset.
   
   I guess the possible solution would be to treat the BeforeFirst as the end 
of epoch and not hard reset only if the end of epoch was actually reached 
through calls to next. I will look into fixing this. 
   Still, the "reset" name kind of suggests it will completely reset the 
iterator and that is not the case if you reach end of epoch. @piiswrong What do 
you think about it? Is it acceptable?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] formath opened a new issue #7738: How to sync files to various machines when submitting job on Yarn

[git]

formath opened a new issue #7738: How to sync files to various machines when 
submitting job on Yarn
URL: https://github.com/apache/incubator-mxnet/issues/7738
 
 
   I can't find any `yarn` example so try myself. But I encounter a problem, 
namely how to sync files to all machines like that of `ssh` mode.
   ```
   ../dmlc-core/tracker/dmlc-submit \
--cluster yarn \
--queue root.default \
--num-workers 2 \
--num-servers 2 \
--worker-cores 1 \
--worker-memory 1g \
--server-cores 1 \
--server-memory 1g \
--ship-libcxx /lib64/ \
--hdfs-tempdir /tmp \
--jobname img-classification \
--archives mxnet.zip \
python ./mxnet/example/image-classification/train_mnist.py --network 
lenet --kv-store dist_sync
   ```
   
   `mxnet.zip` contains
   ```
   python/mxnet
   libmxnet.so
   example/image-classification
   ```
   I know this is problematic like blow. But I don't find the right way. Thanks
   ```
   Traceback (most recent call last):
 File "./train_mnist.py", line 25, in 
   from common import find_mxnet, fit
   ImportError: No module named common
   ```
   
   @tqchen @piiswrong 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] pbertoni commented on issue #2666: Build fail with OpenCV

[git]

pbertoni commented on issue #2666: Build fail with OpenCV
URL: 
https://github.com/apache/incubator-mxnet/issues/2666#issuecomment-327137336
 
 
   Maybe [this](https://github.com/opencv/opencv/issues/9446) build-environment 
related issue could help.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji closed pull request #7737: Use memcpy instead of assigning each individual element

[git]

yaochengji closed pull request #7737: Use memcpy instead of assigning each 
individual element
URL: https://github.com/apache/incubator-mxnet/pull/7737
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji opened a new pull request #7739: Use memcpy instead of assigning each individual element

[git]

yaochengji opened a new pull request #7739: Use memcpy instead of assigning 
each individual element
URL: https://github.com/apache/incubator-mxnet/pull/7739
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] formath closed issue #7738: How to sync files to various machines when submitting job on Yarn

[git]

formath closed issue #7738: How to sync files to various machines when 
submitting job on Yarn
URL: https://github.com/apache/incubator-mxnet/issues/7738
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] Prasad9 opened a new issue #7740: Unable to install MXNet with GPU support for python using pip

[git]

Prasad9 opened a new issue #7740: Unable to install MXNet with GPU support for 
python using pip
URL: https://github.com/apache/incubator-mxnet/issues/7740
 
 
   I am trying to install MXNet for Python using pip having GPU support. I am 
following the instructions mentioned in this 
[documentation](https://mxnet.incubator.apache.org/get_started/install.html). 
My OS is Ubuntu 16.04
   
   After I install MXNet, I run the validation code mentioned in documentation. 
I get the error as stated below.
   
   ```
   >>> import mxnet as mx
   >>> a = mx.nd.ones((2,3), mx.gpu())
   [17:26:19] src/c_api/c_api_ndarray.cc:147: GPU support is disabled. Compile 
MXNet with USE_CUDA=1 to enable GPU support.
   [17:26:19] 
/home/travis/build/dmlc/mxnet-distro/mxnet-build/dmlc-core/include/dmlc/logging.h:308:
 [17:26:19] src/c_api/c_api_ndarray.cc:416: Operator _ones is not implemented 
for GPU.
   
   Stack trace returned 10 entries:
   [bt] (0) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/libmxnet.so(+0xec23c)
 [0x7f298299523c]
   [bt] (1) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/libmxnet.so(+0xde7a52)
 [0x7f2983690a52]
   [bt] (2) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/libmxnet.so(MXImperativeInvoke+0x254)
 [0x7f2983691544]
   [bt] (3) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(ffi_call_unix64+0x4c)
 [0x7f299be918cc]
   [bt] (4) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(ffi_call+0x165)
 [0x7f299be90a15]
   [bt] (5) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(_ctypes_callproc+0x283)
 [0x7f299be88e43]
   [bt] (6) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(+0x96af)
 [0x7f299be806af]
   [bt] (7) 
/home/prasad/miniconda3/envs/myenv/bin/../lib/libpython3.5m.so.1.0(PyObject_Call+0x6a)
 [0x7f299de4aeca]
   [bt] (8) 
/home/prasad/miniconda3/envs/myenv/bin/../lib/libpython3.5m.so.1.0(PyEval_EvalFrameEx+0x36b6)
 [0x7f299df340d6]
   [bt] (9) 
/home/prasad/miniconda3/envs/myenv/bin/../lib/libpython3.5m.so.1.0(+0x15dbd6) 
[0x7f299df3abd6]
   
   Traceback (most recent call last):
 File "", line 1, in 
 File 
"/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/ndarray.py",
 line 1216, in ones
   return _internal._ones(shape=shape, ctx=ctx, dtype=dtype, **kwargs)
 File "", line 15, in _ones
 File 
"/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/_ctypes/ndarray.py",
 line 89, in _imperative_invoke
   c_array(ctypes.c_char_p, [c_str(str(val)) for val in vals])))
 File 
"/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/base.py", 
line 129, in check_call
   raise MXNetError(py_str(_LIB.MXGetLastError()))
   mxnet.base.MXNetError: [17:26:19] src/c_api/c_api_ndarray.cc:416: Operator 
_ones is not implemented for GPU.
   
   Stack trace returned 10 entries:
   [bt] (0) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/libmxnet.so(+0xec23c)
 [0x7f298299523c]
   [bt] (1) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/libmxnet.so(+0xde7a52)
 [0x7f2983690a52]
   [bt] (2) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/site-packages/mxnet/libmxnet.so(MXImperativeInvoke+0x254)
 [0x7f2983691544]
   [bt] (3) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(ffi_call_unix64+0x4c)
 [0x7f299be918cc]
   [bt] (4) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(ffi_call+0x165)
 [0x7f299be90a15]
   [bt] (5) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(_ctypes_callproc+0x283)
 [0x7f299be88e43]
   [bt] (6) 
/home/prasad/miniconda3/envs/myenv/lib/python3.5/lib-dynload/_ctypes.cpython-35m-x86_64-linux-gnu.so(+0x96af)
 [0x7f299be806af]
   [bt] (7) 
/home/prasad/miniconda3/envs/myenv/bin/../lib/libpython3.5m.so.1.0(PyObject_Call+0x6a)
 [0x7f299de4aeca]
   [bt] (8) 
/home/prasad/miniconda3/envs/myenv/bin/../lib/libpython3.5m.so.1.0(PyEval_EvalFrameEx+0x36b6)
 [0x7f299df340d6]
   [bt] (9) 
/home/prasad/miniconda3/envs/myenv/bin/../lib/libpython3.5m.so.1.0(+0x15dbd6) 
[0x7f299df3abd6]
   
   
   ```
   However, if I change the context to CPU, the code runs properly. Please can 
somebody tell me how to install MXNet having GPU support.
   
   Thanks in advance.
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] techzhou commented on issue #7664: index out of bound error when update eval metric

[git]

techzhou commented on issue #7664: index out of bound error when update eval 
metric
URL: 
https://github.com/apache/incubator-mxnet/issues/7664#issuecomment-327160155
 
 
   I have same problem
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger opened a new pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

mseeger opened a new pull request #7741: New operators linalg_syrk, linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741
 
 
   - Works for CPU only for now. We will supply the GPU versions next
   - Added dtype parameters to unit test support code, so we can run numerical 
tests in float64
   
   @asmushetzel
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] javelinjs commented on issue #7716: [Scala] add BucketingModule

[git]

javelinjs commented on issue #7716: [Scala] add BucketingModule
URL: https://github.com/apache/incubator-mxnet/pull/7716#issuecomment-327198112
 
 
   Besides the examples, can we add some unit test cases?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] javelinjs commented on a change in pull request #7716: [Scala] add BucketingModule

[git]

javelinjs commented on a change in pull request #7716: [Scala] add 
BucketingModule
URL: https://github.com/apache/incubator-mxnet/pull/7716#discussion_r137008070
 
 

 ##
 File path: scala-package/examples/scripts/rnn/run_lstm_bucketing.sh
 ##
 @@ -0,0 +1,41 @@
+#!/bin/bash
+
+# Licensed to the Apache Software Foundation (ASF) under one
+# or more contributor license agreements.  See the NOTICE file
+# distributed with this work for additional information
+# regarding copyright ownership.  The ASF licenses this file
+# to you under the Apache License, Version 2.0 (the
+# "License"); you may not use this file except in compliance
+# with the License.  You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing,
+# software distributed under the License is distributed on an
+# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+# KIND, either express or implied.  See the License for the
+# specific language governing permissions and limitations
+# under the License.
+
+
+MXNET_ROOT=$(cd "$(dirname $0)/../../../.."; pwd)
+OS=$(uname)
+if [ "$OS" = "Darwin" ]; then
+  
CLASS_PATH=$MXNET_ROOT/scala-package/assembly/osx-x86_64-gpu/target/*:$MXNET_ROOT/scala-package/examples/target/*:$MXNET_ROOT/scala-package/examples/target/classes/lib/*
+else
+  
CLASS_PATH=$MXNET_ROOT/scala-package/assembly/linux-x86_64-gpu/target/*:$MXNET_ROOT/scala-package/examples/target/*:$MXNET_ROOT/scala-package/examples/target/classes/lib/*
+fi
+
+DATA_TRAIN=$1
+DATA_VAL=$2
+NUM_EPOCH=5
+GPUS="0"
+SAVE_MODEL_PATH=./model/lstm
+
+java -Xmx4G -cp $CLASS_PATH \
+   ml.dmlc.mxnetexamples.rnn.LstmBucketing \
+   --data-train $DATA_TRAIN \
+   --data-val $DATA_VAL \
+   --num-epoch $NUM_EPOCH \
+   --gpus $GPUS \
+   --save-model-path $SAVE_MODEL_PATH
 
 Review comment:
   use 2-space indent
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] javelinjs commented on a change in pull request #7716: [Scala] add BucketingModule

[git]

javelinjs commented on a change in pull request #7716: [Scala] add 
BucketingModule
URL: https://github.com/apache/incubator-mxnet/pull/7716#discussion_r137005931
 
 

 ##
 File path: 
scala-package/core/src/main/scala/ml/dmlc/mxnet/module/BaseModule.scala
 ##
 @@ -22,9 +22,32 @@ import java.io.IOException
 import ml.dmlc.mxnet.optimizer.SGD
 import ml.dmlc.mxnet._
 import org.slf4j.LoggerFactory
-
+import org.slf4j.Logger
 import scala.collection.mutable.ArrayBuffer
 
+object BaseModule {
+  /**
+   * Check that all input names are in symbol's arguments.
+   */
+  @throws(classOf[IllegalArgumentException])
+  def _checkInputNames(symbol: Symbol, names: IndexedSeq[String],
 
 Review comment:
   is it `private`?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] x10000year opened a new issue #7742: NDArray::WaitToWrite() hangs in custom c++ operator. Bug?

[git]

x1year opened a new issue #7742: NDArray::WaitToWrite() hangs in custom c++ 
operator. Bug?
URL: https://github.com/apache/incubator-mxnet/issues/7742
 
 
   I'm writing a custom operator in c++. And I try to construct a NDArray 
object from a blob. I use NDArray because it is more dynamic and easier to use 
than mshadow::Tensor. However, I found that a NDAarray object constructed from 
an output blob hangs on call to WaitToWrite(). Is this a bug?
   
   The code is as following:
   
 void Forward(const OpContext &ctx,
 const std::vector &in_data,
 const std::vector &req,
 const std::vector &out_data,
 const std::vector &aux_args) override {
   CHECK(req[0] == kWriteTo);
   NDArray x(out_data[0], out_data[0].dev_id());
   x.WaitToWrite();  // hangs here
   ...
   x.SyncCopyFromCPU(...);  // which calls WaitToWrite() inside
   }
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] cjolivier01 commented on issue #7577: Sparse operators for unary and binary elemwise NDArray operators.

[git]

cjolivier01 commented on issue #7577: Sparse operators for unary and binary 
elemwise NDArray operators.
URL: https://github.com/apache/incubator-mxnet/pull/7577#issuecomment-326788688
 
 
   In my build_test branch I have several iterations. Some built, just finding
   one that's closest to the proper intent that msvc won't choke on.
   
   On Sun, Sep 3, 2017 at 12:14 AM Chris Olivier  wrote:
   
   > On Windows?  Mine passed on the build server AMI. I can get several
   > permutations to pass. I will get it working on Tuesday.
   >
   >
   > On Sat, Sep 2, 2017 at 11:31 PM Haibin Lin 
   > wrote:
   >
   >> I was trying to build your branch on Ubuntu DL AMI but make failed with
   >> the same errors that appeared in CI
   >>
   >> ?
   >> You are receiving this because you authored the thread.
   >> Reply to this email directly, view it on GitHub
   >> 
,
   >> or mute the thread
   >> 

   >> .
   >>
   >
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #7577: Sparse operators for unary and binary elemwise NDArray operators.

[git]

eric-haibin-lin commented on issue #7577: Sparse operators for unary and binary 
elemwise NDArray operators.
URL: https://github.com/apache/incubator-mxnet/pull/7577#issuecomment-326787206
 
 
   I was trying to build your branch on Ubuntu DL AMI but `make` failed with 
the same errors that appeared in CI 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] cjolivier01 commented on issue #7577: Sparse operators for unary and binary elemwise NDArray operators.

[git]

cjolivier01 commented on issue #7577: Sparse operators for unary and binary 
elemwise NDArray operators.
URL: https://github.com/apache/incubator-mxnet/pull/7577#issuecomment-326788610
 
 
   On Windows?  Mine passed on the build server AMI. I can get several
   permutations to pass. I will get it working on Tuesday.
   
   
   On Sat, Sep 2, 2017 at 11:31 PM Haibin Lin  wrote:
   
   > I was trying to build your branch on Ubuntu DL AMI but make failed with
   > the same errors that appeared in CI
   >
   > ?
   > You are receiving this because you authored the thread.
   > Reply to this email directly, view it on GitHub
   > 
,
   > or mute the thread
   > 

   > .
   >
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #7577: Sparse operators for unary and binary elemwise NDArray operators.

[git]

eric-haibin-lin commented on issue #7577: Sparse operators for unary and binary 
elemwise NDArray operators.
URL: https://github.com/apache/incubator-mxnet/pull/7577#issuecomment-326815376
 
 
   Thanks. It's not on windows.. I did a clean `clone -b tsparse_prmaster 
--recursive` on `Deep Learning AMI Ubuntu Version-1-0` 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] asmushetzel commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

asmushetzel commented on a change in pull request #7741: New operators 
linalg_syrk, linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137020459
 
 

 ##
 File path: src/operator/tensor/la_op.cc
 ##
 @@ -95,33 +96,33 @@ NNVM_REGISTER_OP(_backward_linalg_gemm)
 NNVM_REGISTER_OP(_linalg_gemm2)
 .add_alias("linalg_gemm2")
 .describe(R"code(Performs general matrix multiplication.
-Input are two tensors *A*, *B* each of dimension *n >= 2* and each
-having the same shape on the leading *n-2* dimensions. For every *n-2* 
dimensional index *i* let
-*A*\ :sub:`i`\ , *B*\ :sub:`i`\  be the matrices given by the last *2* 
dimensions.
-The operator performs the BLAS3 function *gemm* (restricted to two arguments)
+Input are tensors *A*, *B*, each of dimension *n >= 2* and having the same 
shape
+on the leading *n-2* dimensions.
+
+If *n=2*, the BLAS3 function *gemm* is performed:
 
-   *out*\ :sub:`i` = *alpha* \* *op*\ (*A*\ :sub:`i`\ ) \* *op*\ (*B*\ 
:sub:`i`\ )
+   *out* = *alpha* \* *op*\ (*A*) \* *op*\ (*B*)
 
-on all such pairs of matrices. Here *alpha* is a scalar operator parameter and 
*op()* is either
-the identity or the matrix transposition.
+Here *alpha* is a scalar parameter and *op()* is either the identity or the 
matrix
+transposition (depending on *transpose_a*, *transpose_b*).
 
-In case of *n=2*, a single *gemm* function is performed on the matrices *A*, 
*B*.
+If *n>2*, *gemm* is performed separately on the trailing two dimensions for 
all inputs
+(batch mode).
 
-.. note:: The operator does only support float32 and float64 data types and 
provides
-  proper backward gradients.
+.. note:: The operator supports float32 and float64 data types only.
 
 Examples::
 
// Single matrix multiply
A = [[1.0, 1.0], [1.0, 1.0]]
B = [[1.0, 1.0], [1.0, 1.0], [1.0, 1.0]]
-   gemm2(A, B, transpose_b = 1, alpha = 2.0)
 
 Review comment:
   Apparently some general renaming has happened recently for namespace 
reasons. All documentation now uses function names without linalg_prefix.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] JeanKossaifi opened a new pull request #7743: Minor update: correct badge in README

[git]

JeanKossaifi opened a new pull request #7743: Minor update: correct badge in 
README
URL: https://github.com/apache/incubator-mxnet/pull/7743
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] asmushetzel commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

asmushetzel commented on a change in pull request #7741: New operators 
linalg_syrk, linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137020947
 
 

 ##
 File path: src/operator/tensor/la_op.cc
 ##
 @@ -251,42 +253,42 @@ NNVM_REGISTER_OP(_backward_linalg_potri)
 .set_attr("TIsBackward", true)
 .set_attr("FCompute", LaOpBackward);
 
-NNVM_REGISTER_OP(_linalg_trmm)
+NNVM_REGISTER_OP(linalg_trmm)
 
 Review comment:
   All these operators are now registered with an initial underscore followed 
by an alias for backward compatibility. 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] asmushetzel commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

asmushetzel commented on a change in pull request #7741: New operators 
linalg_syrk, linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137021425
 
 

 ##
 File path: src/operator/tensor/la_op.cc
 ##
 @@ -420,5 +421,134 @@ NNVM_REGISTER_OP(_backward_linalg_sumlogdiag)
 .set_attr("TIsBackward", true)
 .set_attr("FCompute", LaOpBackward);
 
+NNVM_REGISTER_OP(linalg_syrk)
 
 Review comment:
   Initial underscore followed by alias (as well as in the other new operators)
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji closed pull request #7739: Use memcpy instead of assigning each individual element

[git]

yaochengji closed pull request #7739: Use memcpy instead of assigning each 
individual element
URL: https://github.com/apache/incubator-mxnet/pull/7739
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #7740: Unable to install MXNet with GPU support for python using pip

[git]

szha commented on issue #7740: Unable to install MXNet with GPU support for 
python using pip
URL: 
https://github.com/apache/incubator-mxnet/issues/7740#issuecomment-327213804
 
 
   Could you paste the command you used when installing from pip and the output 
of that command?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] asmushetzel commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

asmushetzel commented on a change in pull request #7741: New operators 
linalg_syrk, linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137025280
 
 

 ##
 File path: src/operator/tensor/la_op.h
 ##
 @@ -200,9 +216,7 @@ inline bool LaReduceShape(const nnvm::NodeAttrs& attrs,
   CHECK_EQ(in_attrs->size(), 1);
   CHECK_EQ(out_attrs->size(), 1);
   const int ndim((*in_attrs)[0].ndim());
-  if ( ndim < dim ) {
- return false;
-  }
+  CHECK_GE(ndim, dim) << "Shape of input has too few dimensions";
 
 Review comment:
   Please leave original code. We may have ndim < dim in the case where the 
input shape has not yet been defined at all (but may be defined later). Means 
at this point in time we can't do any inference but not necessary that we can't 
ever (in a subsequent pass).  
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #7735: My jupyter notebook uses old mxnet version 0.9.5 whereas pycharm debugger uses 0.11.1...

[git]

szha commented on issue #7735: My jupyter notebook uses old mxnet version 0.9.5 
whereas pycharm debugger uses 0.11.1...
URL: 
https://github.com/apache/incubator-mxnet/issues/7735#issuecomment-327216932
 
 
   You might have more than one copy of python and your ipython/Jupyer has old 
mxnet in its site packages and it wasn't upgraded. You can open a notebook, 
import mxnet, and then print mxnet to find where the old package is. Note down 
the location, find the pip command in the bin folder at the same level as the 
site-packages you found from last step, and explicitly use the absolute path of 
the pip you found for installation and upgrade.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7743: Minor update: correct badge in README

[git]

szha commented on a change in pull request #7743: Minor update: correct badge 
in README
URL: https://github.com/apache/incubator-mxnet/pull/7743#discussion_r137028577
 
 

 ##
 File path: README.md
 ##
 @@ -1,7 +1,7 @@
 Apache MXNet (incubating) for Deep Learning
 =
 
-[![Build 
Status](https://travis-ci.org/dmlc/mxnet.svg?branch=master)](https://travis-ci.org/dmlc/mxnet)
+[![Build 
Status](https://travis-ci.org/apache/incubator-mxnet.svg?branch=master)](https://travis-ci.org/apache/incubator-mxnet)
 
 Review comment:
   Travis is disabled for now, and we are using Jenkins for CI.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] goodtogood commented on issue #7725: accuracy of cpp example is a constant value when training, no matter how many epochs trained!

[git]

goodtogood commented on issue #7725: accuracy of cpp example is a constant 
value when training, no matter how many epochs trained!
URL: 
https://github.com/apache/incubator-mxnet/issues/7725#issuecomment-327218807
 
 
   @piiswrong @mli 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

mseeger commented on a change in pull request #7741: New operators linalg_syrk, 
linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137030821
 
 

 ##
 File path: src/operator/tensor/la_op.cc
 ##
 @@ -251,42 +253,42 @@ NNVM_REGISTER_OP(_backward_linalg_potri)
 .set_attr("TIsBackward", true)
 .set_attr("FCompute", LaOpBackward);
 
-NNVM_REGISTER_OP(_linalg_trmm)
+NNVM_REGISTER_OP(linalg_trmm)
 
 Review comment:
   OK, I will do this
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

mseeger commented on a change in pull request #7741: New operators linalg_syrk, 
linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137030737
 
 

 ##
 File path: src/operator/tensor/la_op.cc
 ##
 @@ -95,33 +96,33 @@ NNVM_REGISTER_OP(_backward_linalg_gemm)
 NNVM_REGISTER_OP(_linalg_gemm2)
 .add_alias("linalg_gemm2")
 .describe(R"code(Performs general matrix multiplication.
-Input are two tensors *A*, *B* each of dimension *n >= 2* and each
-having the same shape on the leading *n-2* dimensions. For every *n-2* 
dimensional index *i* let
-*A*\ :sub:`i`\ , *B*\ :sub:`i`\  be the matrices given by the last *2* 
dimensions.
-The operator performs the BLAS3 function *gemm* (restricted to two arguments)
+Input are tensors *A*, *B*, each of dimension *n >= 2* and having the same 
shape
+on the leading *n-2* dimensions.
+
+If *n=2*, the BLAS3 function *gemm* is performed:
 
-   *out*\ :sub:`i` = *alpha* \* *op*\ (*A*\ :sub:`i`\ ) \* *op*\ (*B*\ 
:sub:`i`\ )
+   *out* = *alpha* \* *op*\ (*A*) \* *op*\ (*B*)
 
-on all such pairs of matrices. Here *alpha* is a scalar operator parameter and 
*op()* is either
-the identity or the matrix transposition.
+Here *alpha* is a scalar parameter and *op()* is either the identity or the 
matrix
+transposition (depending on *transpose_a*, *transpose_b*).
 
-In case of *n=2*, a single *gemm* function is performed on the matrices *A*, 
*B*.
+If *n>2*, *gemm* is performed separately on the trailing two dimensions for 
all inputs
+(batch mode).
 
-.. note:: The operator does only support float32 and float64 data types and 
provides
-  proper backward gradients.
+.. note:: The operator supports float32 and float64 data types only.
 
 Examples::
 
// Single matrix multiply
A = [[1.0, 1.0], [1.0, 1.0]]
B = [[1.0, 1.0], [1.0, 1.0], [1.0, 1.0]]
-   gemm2(A, B, transpose_b = 1, alpha = 2.0)
 
 Review comment:
   OK, I will change the docstrings
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mseeger commented on a change in pull request #7741: New operators linalg_syrk, linalg_gelqf

[git]

mseeger commented on a change in pull request #7741: New operators linalg_syrk, 
linalg_gelqf
URL: https://github.com/apache/incubator-mxnet/pull/7741#discussion_r137032345
 
 

 ##
 File path: src/operator/tensor/la_op.cc
 ##
 @@ -420,5 +421,134 @@ NNVM_REGISTER_OP(_backward_linalg_sumlogdiag)
 .set_attr("TIsBackward", true)
 .set_attr("FCompute", LaOpBackward);
 
+NNVM_REGISTER_OP(linalg_syrk)
 
 Review comment:
   OK, will do this
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #7390: adding ranking metrics (precision/recall) at position K.

[git]

eric-haibin-lin commented on issue #7390: adding ranking metrics 
(precision/recall) at position K. 
URL: https://github.com/apache/incubator-mxnet/pull/7390#issuecomment-327228807
 
 
   @piiswrong merge? 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #7390: adding ranking metrics (precision/recall) at position K.

[git]

szha commented on issue #7390: adding ranking metrics (precision/recall) at 
position K. 
URL: https://github.com/apache/incubator-mxnet/pull/7390#issuecomment-327234630
 
 
   Honestly I'd rather see the logic for confusion matrix entries in one place 
instead of 6.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on issue #7390: adding ranking metrics (precision/recall) at position K.

[git]

eric-haibin-lin commented on issue #7390: adding ranking metrics 
(precision/recall) at position K. 
URL: https://github.com/apache/incubator-mxnet/pull/7390#issuecomment-327228807
 
 
   @piiswrong merge? 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] piiswrong commented on a change in pull request #7732: Update optimizer.py

[git]

piiswrong commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137051751
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   ?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7732: Update optimizer.py

[git]

szha commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137053173
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   self.states can be length 2 tuple so this is a bug fix.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] larroy commented on issue #7416: update submoules with android fixes

[git]

larroy commented on issue #7416: update submoules with android fixes
URL: https://github.com/apache/incubator-mxnet/pull/7416#issuecomment-327240450
 
 
   done
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] piiswrong closed pull request #7730: display type string instead of type code in error message

[git]

piiswrong closed pull request #7730: display type string instead of type code 
in error message
URL: https://github.com/apache/incubator-mxnet/pull/7730
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: display type string instead of type code in error message (#7730)

[jxie]

This is an automated email from the ASF dual-hosted git repository.

jxie pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new b3ad753  display type string instead of type code in error message 
(#7730)
b3ad753 is described below

commit b3ad75311fb7012c17cde3c56425af580db6934a
Author: Sheng Zha 
AuthorDate: Tue Sep 5 10:09:30 2017 -0700

display type string instead of type code in error message (#7730)
---
 src/operator/activation-inl.h |  4 +---
 src/operator/batch_norm-inl.h |  8 ++--
 src/operator/batch_norm_v1-inl.h  |  8 ++--
 src/operator/contrib/count_sketch-inl.h   |  4 +---
 src/operator/contrib/deformable_convolution-inl.h |  4 +---
 src/operator/contrib/fft-inl.h|  4 +---
 src/operator/contrib/ifft-inl.h   |  4 +---
 src/operator/convolution-inl.h|  4 +---
 src/operator/convolution_v1-inl.h |  4 +---
 src/operator/deconvolution-inl.h  |  4 +---
 src/operator/lrn-inl.h|  4 +---
 src/operator/operator_common.h| 13 +
 src/operator/rnn-inl.h|  4 +---
 src/operator/sequence_last-inl.h  |  5 +
 src/operator/sequence_mask-inl.h  |  5 +
 src/operator/sequence_reverse-inl.h   |  5 +
 src/operator/softmax_output-inl.h |  4 +---
 src/operator/svm_output-inl.h |  4 +---
 src/operator/upsampling-inl.h |  4 +---
 19 files changed, 33 insertions(+), 63 deletions(-)

diff --git a/src/operator/activation-inl.h b/src/operator/activation-inl.h
index 6e6222b..8b1a229 100644
--- a/src/operator/activation-inl.h
+++ b/src/operator/activation-inl.h
@@ -137,9 +137,7 @@ class ActivationProp : public OperatorProperty {
   if ((*in_type)[i] == -1) {
   (*in_type)[i] = dtype;
   } else {
-CHECK_EQ((*in_type)[i], dtype) << "This layer requires uniform type. "
-   << "Expected " << dtype << " v.s. given 
"
-   << (*in_type)[i] << " at " << 
ListArguments()[i];
+UNIFORM_TYPE_CHECK((*in_type)[i], dtype, ListArguments()[i]);
   }
 }
 out_type->clear();
diff --git a/src/operator/batch_norm-inl.h b/src/operator/batch_norm-inl.h
index b53acb4..461f702 100644
--- a/src/operator/batch_norm-inl.h
+++ b/src/operator/batch_norm-inl.h
@@ -274,16 +274,12 @@ class BatchNormProp : public OperatorProperty {
   if ((*in_type)[i] == -1) {
 (*in_type)[i] = dtype_param;
   } else {
-CHECK_EQ((*in_type)[i], dtype_param) << "This layer requires uniform 
type. "
- << "Expected " << dtype_param << 
" v.s. given "
- << (*in_type)[i] << " at " << 
ListArguments()[i];
+UNIFORM_TYPE_CHECK((*in_type)[i], dtype_param, ListArguments()[i]);
   }
 }
 for (index_t i = 0; i < aux_type->size(); ++i) {
   if ((*aux_type)[i] != -1) {
-CHECK_EQ((*aux_type)[i], dtype_param) << "This layer requires uniform 
type. "
-  << "Expected " << dtype_param << 
" v.s. given "
-  << (*aux_type)[i] << " at " << 
ListArguments()[i];
+UNIFORM_TYPE_CHECK((*aux_type)[i], dtype_param, ListArguments()[i]);
   }
 }
 const size_t n_aux = this->ListAuxiliaryStates().size();
diff --git a/src/operator/batch_norm_v1-inl.h b/src/operator/batch_norm_v1-inl.h
index 092c482..ebfc469 100644
--- a/src/operator/batch_norm_v1-inl.h
+++ b/src/operator/batch_norm_v1-inl.h
@@ -289,16 +289,12 @@ class BatchNormV1Prop : public OperatorProperty {
   if ((*in_type)[i] == -1) {
 (*in_type)[i] = dtype_param;
   } else {
-CHECK_EQ((*in_type)[i], dtype_param) << "This layer requires uniform 
type. "
- << "Expected " << dtype_param << 
" v.s. given "
- << (*in_type)[i] << " at " << 
ListArguments()[i];
+UNIFORM_TYPE_CHECK((*in_type)[i], dtype_param, ListArguments()[i]);
   }
 }
 for (index_t i = 0; i < aux_type->size(); ++i) {
   if ((*aux_type)[i] != -1) {
-CHECK_EQ((*aux_type)[i], dtype_param) << "This layer requires uniform 
type. "
-  << "Expected " << dtype_param << 
" v.s. given "
-  << (*aux_type)[i] << " at " << 
ListArguments()[i];
+UNIFORM_TYPE_CHECK((*aux_type)[i], dtype_param, ListArguments()[i]);
   }
 }
 int n_aux = this->ListAuxiliaryStates().size();
diff --git a/src/operator/contrib/count_sketch-inl

[GitHub] piiswrong commented on issue #7742: NDArray::WaitToWrite() hangs in custom c++ operator. Bug?

[git]

piiswrong commented on issue #7742: NDArray::WaitToWrite() hangs in custom c++ 
operator. Bug?
URL: 
https://github.com/apache/incubator-mxnet/issues/7742#issuecomment-327243259
 
 
   You can't use NDArray inside operators
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7732: Update optimizer.py

[git]

szha commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137053173
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   `self.states` (i.e. the optimizer's states) can be length 2 tuple so this is 
a bug fix. Here the `states` is the deserialized form, which can either be 
tuple or dict after this change. See line 998.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: update submoules with android fixes (#7416)

[jxie]

This is an automated email from the ASF dual-hosted git repository.

jxie pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new d6cc2a8  update submoules with android fixes (#7416)
d6cc2a8 is described below

commit d6cc2a877fbaa45a6474c6d625da6ed41af8cc5f
Author: Pedro Larroy <928489+lar...@users.noreply.github.com>
AuthorDate: Tue Sep 5 19:26:23 2017 +0200

update submoules with android fixes (#7416)

* [Android] Address different return type from LAPACK functions when using 
OpenBlas

* update mshadow
---
 mshadow | 2 +-
 src/operator/c_lapack_api.h | 9 +++--
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/mshadow b/mshadow
index 51c4043..15d2533 16
--- a/mshadow
+++ b/mshadow
@@ -1 +1 @@
-Subproject commit 51c40430711bbea807efd256ee2dbacce0081276
+Subproject commit 15d2533f3a92e7cdbcd081114f8e392f4b7e5c67
diff --git a/src/operator/c_lapack_api.h b/src/operator/c_lapack_api.h
index 96a9b3a..81f07ef 100644
--- a/src/operator/c_lapack_api.h
+++ b/src/operator/c_lapack_api.h
@@ -73,8 +73,13 @@ using namespace mshadow;
 extern "C" {
 
   // Fortran signatures
-  #define MXNET_LAPACK_FSIGNATURE1(func, dtype) \
-void func##_(char *uplo, int *n, dtype *a, int *lda, int *info);
+  #ifdef __ANDROID__
+#define MXNET_LAPACK_FSIGNATURE1(func, dtype) \
+  int func##_(char* uplo, int* n, dtype* a, int* lda, int *info);
+  #else
+#define MXNET_LAPACK_FSIGNATURE1(func, dtype) \
+  void func##_(char* uplo, int* n, dtype* a, int* lda, int *info);
+  #endif
 
   MXNET_LAPACK_FSIGNATURE1(spotrf, float)
   MXNET_LAPACK_FSIGNATURE1(dpotrf, double)

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" '].

[GitHub] piiswrong closed pull request #7416: update submoules with android fixes

[git]

piiswrong closed pull request #7416: update submoules with android fixes
URL: https://github.com/apache/incubator-mxnet/pull/7416
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #7390: adding ranking metrics (precision/recall) at position K.

[git]

eric-haibin-lin commented on a change in pull request #7390: adding ranking 
metrics (precision/recall) at position K. 
URL: https://github.com/apache/incubator-mxnet/pull/7390#discussion_r137061206
 
 

 ##
 File path: python/mxnet/metric.py
 ##
 @@ -568,6 +568,147 @@ def update(self, labels, preds):
 self.sum_metric += f1_score
 self.num_inst += 1
 
+@register
+@alias('top_k_precision')
+class TopKPrecision(EvalMetric):
+"""Computes top k precision metric.
+top k differs from regular precision in that the score is only
+computed for the top k predictions. "correct" or "wrong" entries
+outside the top k are ignored
+Parameters
+--
+top_k : int
+Whether targets are in top k predictions.
+name : str
+Name of this metric instance for display.
+output_names : list of str, or None
+Name of predictions that should be used when updating with update_dict.
+By default include all predictions.
+label_names : list of str, or None
+Name of labels that should be used when updating with update_dict.
+By default include all labels.
+
+Examples
+
+>>>ytrue = [[1.,0.,1.,0.],[0.,1.,1.,0.]]
+>>>ytrue = mx.nd.array(ytrue)
+>>>yhat = [[0.4,0.8,0.1,0.1],[0.4,0.8,0.8,0.4]]
+>>>yhat = mx.nd.array(yhat)
+>>>pre = mx.metric.create('top_k_precision',top_k=2)
+>>>pre.update(preds = [yhat], labels = [ytrue])
+>>>print pre.get()[1]
+>>> 0.75
+
+"""
+
+def __init__(self, top_k=1, name='top_k_precision',
+ output_names=None, label_names=None):
+super(TopKPrecision, self).__init__(
+name, top_k=top_k,
+output_names=output_names, label_names=label_names)
+self.top_k = top_k
+
+
+def update(self, labels, preds):
+"""Updates the internal evaluation result.
+Parameters
+--
+labels : list of `NDArray`
+The labels of the data. (binary)
+preds : list of `NDArray`
+Predicted values. (float)
+
+Returns:
+
+The precision at K (float)
+"""
+check_label_shapes(labels, preds)
+
+for label, pred_label in zip(labels, preds):
+assert(len(pred_label.shape) <= 2), 'Predictions should be no more 
than 2 dims'
+pred_label = 
numpy.argsort(-pred_label.asnumpy().astype('float32'), axis=1)
+label = label.asnumpy().astype('int32')
+check_label_shapes(label, pred_label)
+num_samples = pred_label.shape[0]
+local_precision = 0.0
+for s in range(num_samples):
+truepos = set(numpy.ravel(numpy.argwhere(label[s, :] == 1)))
 
 Review comment:
   The logics to compute `truepos` and `predpos` are redundant in the two 
classes. Could you refactor it as a free function so that the two classes can 
call, or let the two classes inherit from an abstract class that has such a 
method? 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] piiswrong commented on a change in pull request #7732: Update optimizer.py

[git]

piiswrong commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137062908
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   But states are previously a dictionary. This is not backward compatible
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] piiswrong commented on a change in pull request #7732: Update optimizer.py

[git]

piiswrong commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137063049
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   why would it be a tuple?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7732: Update optimizer.py

[git]

szha commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137064283
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   I missed this myself, it's here: 
https://github.com/apache/incubator-mxnet/blob/master/python/mxnet/optimizer.py#L375
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7732: Update optimizer.py

[git]

szha commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137065155
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   List then?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] Roshrini commented on issue #7600: benchmark.py modified

[git]

Roshrini commented on issue #7600: benchmark.py modified
URL: https://github.com/apache/incubator-mxnet/pull/7600#issuecomment-327253171
 
 
   @nswamy 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on a change in pull request #7732: Update optimizer.py

[git]

szha commented on a change in pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#discussion_r137077424
 
 

 ##
 File path: python/mxnet/optimizer.py
 ##
 @@ -979,8 +979,9 @@ def sync_state_context(self, state, context):
 def set_states(self, states):
 """Sets updater states."""
 states = pickle.loads(states)
-if isinstance(states, tuple) and len(states) == 2:
-self.states, self.optimizer = states
+if isinstance(states, dict):
 
 Review comment:
   I was confused. `self.states` is actually the dictionary in Updater, so the 
optimizer's tuple shouldn't be a problem.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] ashokei opened a new pull request #7744: multinode dist_sync fails with mklml

[git]

ashokei opened a new pull request #7744: multinode dist_sync fails with mklml
URL: https://github.com/apache/incubator-mxnet/pull/7744
 
 
   multinode dist_sync fails with mklml, this patch uses utility function in 
tensor_blob to handle it correctly.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha commented on issue #7732: Update optimizer.py

[git]

szha commented on issue #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732#issuecomment-327264295
 
 
   Closing for now. A more general refactor on entity serialization in mxnet 
may be needed.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] szha closed pull request #7732: Update optimizer.py

[git]

szha closed pull request #7732: Update optimizer.py
URL: https://github.com/apache/incubator-mxnet/pull/7732
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] piiswrong commented on issue #7744: multinode dist_sync fails with mklml

[git]

piiswrong commented on issue #7744: multinode dist_sync fails with mklml
URL: https://github.com/apache/incubator-mxnet/pull/7744#issuecomment-327264531
 
 
   Could you also fix it for 
https://github.com/ashokei/incubator-mxnet/blob/1e03a82dc72fc38f47bbffe010f1c1fe2edb381e/src/kvstore/kvstore_dist.h#L131
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch v0.11.0 updated: Update NEWS.md

[nswamy]

This is an automated email from the ASF dual-hosted git repository.

nswamy pushed a commit to branch v0.11.0
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/v0.11.0 by this push:
 new 1f5863c  Update NEWS.md
1f5863c is described below

commit 1f5863c28c2b33d75a13aafdf94a12f0aabd0268
Author: Naveen Swamy 
AuthorDate: Tue Sep 5 11:36:08 2017 -0700

Update NEWS.md

0.11 release
---
 NEWS.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/NEWS.md b/NEWS.md
index da2ba04..85d6c8d 100644
--- a/NEWS.md
+++ b/NEWS.md
@@ -1,6 +1,6 @@
 MXNet Change Log
 
-## 0.11.0-rc3
+## 0.11.0
 ### - Major Features
   - [Apple Core ML model 
converter](https://github.com/apache/incubator-mxnet/blob/master/tools/coreml/README.md)
   - Support for [Keras v1.2.2](https://github.com/dmlc/keras/wiki/Installation)

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" '].

[GitHub] piiswrong closed pull request #7699: Fix bug in rpn.py

[git]

piiswrong closed pull request #7699: Fix bug in rpn.py
URL: https://github.com/apache/incubator-mxnet/pull/7699
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: Update rpn.py (#7699)

[jxie]

This is an automated email from the ASF dual-hosted git repository.

jxie pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new f39e3ed  Update rpn.py (#7699)
f39e3ed is described below

commit f39e3ed2c987f40b00b3f7b0b337abb3c9146a66
Author: hyxie <493063...@qq.com>
AuthorDate: Wed Sep 6 02:37:17 2017 +0800

Update rpn.py (#7699)
---
 example/rcnn/rcnn/io/rpn.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/example/rcnn/rcnn/io/rpn.py b/example/rcnn/rcnn/io/rpn.py
index 20cd1ce..59dd615 100644
--- a/example/rcnn/rcnn/io/rpn.py
+++ b/example/rcnn/rcnn/io/rpn.py
@@ -188,7 +188,7 @@ def assign_anchor(feat_shape, gt_boxes, im_info, 
feat_stride=16,
 fg_inds = np.where(labels == 1)[0]
 if len(fg_inds) > num_fg:
 disable_inds = npr.choice(fg_inds, size=(len(fg_inds) - num_fg), 
replace=False)
-if logger.level == logging.INFO:
+if logger.level == logging.DEBUG:
 disable_inds = fg_inds[:(len(fg_inds) - num_fg)]
 labels[disable_inds] = -1
 
@@ -197,7 +197,7 @@ def assign_anchor(feat_shape, gt_boxes, im_info, 
feat_stride=16,
 bg_inds = np.where(labels == 0)[0]
 if len(bg_inds) > num_bg:
 disable_inds = npr.choice(bg_inds, size=(len(bg_inds) - num_bg), 
replace=False)
-if logger.level == logging.INFO:
+if logger.level == logging.DEBUG:
 disable_inds = bg_inds[:(len(bg_inds) - num_bg)]
 labels[disable_inds] = -1
 

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" '].

[incubator-mxnet] branch v0.11.0 updated: Update README.md

[nswamy]

This is an automated email from the ASF dual-hosted git repository.

nswamy pushed a commit to branch v0.11.0
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/v0.11.0 by this push:
 new a5edbf9  Update README.md
a5edbf9 is described below

commit a5edbf94094581ee27157eae4f2113115a3994e7
Author: Naveen Swamy 
AuthorDate: Tue Sep 5 11:40:19 2017 -0700

Update README.md
---
 README.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/README.md b/README.md
index d55be79..96f6af3 100644
--- a/README.md
+++ b/README.md
@@ -22,7 +22,7 @@ deep learning systems, and interesting insights of DL systems 
for hackers.
 
 What's New
 --
-* [Version 0.11.0.rc3 
Release](https://github.com/apache/incubator-mxnet/releases/tag/0.11.0.rc3) - 
MXNet 0.11.0.rc3 Release.
+* [Version 0.11.0 
Release](https://github.com/apache/incubator-mxnet/releases/tag/0.11.0) - MXNet 
0.11.0 Release.
 * [Apache Incubator](http://incubator.apache.org/projects/mxnet.html) - We are 
now an Apache Incubator project.
 * [Version 0.10.0 Release](https://github.com/dmlc/mxnet/releases/tag/v0.10.0) 
- MXNet 0.10.0 Release.
 * [Version 0.9.3 Release](./docs/architecture/release_note_0_9.md) - First 0.9 
official release.

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" '].

[GitHub] gokererdogan opened a new issue #7745: Bug in DetRandomCropAug

[git]

gokererdogan opened a new issue #7745: Bug in DetRandomCropAug
URL: https://github.com/apache/incubator-mxnet/issues/7745
 
 
   For bugs or installation issues, please provide the following information.
   The more information you provide, the more likely people will be able to 
help you.
   
   ## Environment info
   Operating System: Ubuntu 16.04
   
   Compiler: gcc 5.4
   
   Package used: Python
   
   MXNet commit hash (`git rev-parse HEAD`): 
ada6d4e0bbfb6a244a868c8ef6edf40529dd996d
   
   If you are using python package, please provide
   
   Python version and distribution: 3.5.2 (from Ubuntu repo)
   
   ## Error Message:
   ```
   File "/mxnet/python/mxnet/image/detection.py", line 750, in next
   data, label = self.augmentation_transform(data, label)
 File "/mxnet/python/mxnet/image/detection.py", line 772, in 
augmentation_transform
   data, label = aug(data, label)
 File "/mxnet/python/mxnet/image/detection.py", line 104, in __call__
   return self.aug_list[0](src, label)
 File "/mxnet/python/mxnet/image/detection.py", line 188, in __call__
   crop = self._random_crop_proposal(label, src.shape[0], src.shape[1])
 File "/mxnet/python/mxnet/image/detection.py", line 299, in 
_random_crop_proposal
   if self._check_satisfy_constraints(label, x, y, x + w, y + h, width, 
height):
 File "/mxnet/python/mxnet/image/detection.py", line 229, in 
_check_satisfy_constraints
   coverages = self._calculate_areas(intersects) / object_areas
   ValueError: operands could not be broadcast together with shapes (2,) (3,) 
   ```
   
   ## Minimum reproducible example
   In `DetRandomCropAug._check_satisfy_constraints`, if after cropping, the 
number of "valid" objects change, the line below throws the above exception. 
   ```
   coverages = self._calculate_areas(intersects) / object_areas
   ```
   I believe this line should read
   ```
   coverages = self._calculate_areas(intersects) / object_areas[valid_objects]
   ```
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] sandeep-krishnamurthy commented on issue #7678: Clean amalgamation ws and skip failing test

[git]

sandeep-krishnamurthy commented on issue #7678: Clean amalgamation ws and skip 
failing test
URL: https://github.com/apache/incubator-mxnet/pull/7678#issuecomment-327270194
 
 
   Remove logic to skipping the failing test case. Thanks to Aston for fixing 
it.
   This PR now contains only fix for amalgamation min CI.
   @madjam - Can we merge this?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

svn commit: r21468 - in /dev/incubator/mxnet/0.11.0: ./ apache-mxnet-src-0.11.0-incubating.tar.gz apache-mxnet-src-0.11.0-incubating.tar.gz.asc apache-mxnet-src-0.11.0-incubating.tar.gz.md5 apache-mxn

[nswamy]

Author: nswamy
Date: Tue Sep  5 19:02:41 2017
New Revision: 21468

Log:
Add mxnet 0.11.0

Added:
dev/incubator/mxnet/0.11.0/
dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz   
(with props)
dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.asc
dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.md5
dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.sha512

Added: dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz
==
Binary file - no diff available.

Propchange: dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz
--
svn:mime-type = application/octet-stream

Added: dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.asc
==
--- dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.asc 
(added)
+++ dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.asc 
Tue Sep  5 19:02:41 2017
@@ -0,0 +1,16 @@
+-BEGIN PGP SIGNATURE-
+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+=oOMj
+-END PGP SIGNATURE-

Added: dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.md5
==
--- dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.md5 
(added)
+++ dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.md5 
Tue Sep  5 19:02:41 2017
@@ -0,0 +1 @@
+MD5 (apache-mxnet-src-0.11.0-incubating.tar.gz) = 
7e942cf9a1be3fbd180f5cab56d71f68

Added: 
dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.sha512
==
--- dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.sha512 
(added)
+++ dev/incubator/mxnet/0.11.0/apache-mxnet-src-0.11.0-incubating.tar.gz.sha512 
Tue Sep  5 19:02:41 2017
@@ -0,0 +1 @@
+0c4c882bfd185acd87dab1a8224399398a7e428e497c875b6f5722e1942bda61bc150b3acc5385b2143f29feeab47e833fb8e3779f3077a203052be47a89240d
  apache-mxnet-src-0.11.0-incubating.tar.gz

svn commit: r21471 - in /release/incubator/mxnet: ./ KEYS

[nswamy]

Author: nswamy
Date: Tue Sep  5 19:20:45 2017
New Revision: 21471

Log:
Add mxnet folder

Added:
release/incubator/mxnet/
release/incubator/mxnet/KEYS

Added: release/incubator/mxnet/KEYS
==
--- release/incubator/mxnet/KEYS (added)
+++ release/incubator/mxnet/KEYS Tue Sep  5 19:20:45 2017
@@ -0,0 +1,191 @@
+This file contains the PGP keys of various developers.
+Please don't use them for email unless you have to. Their main
+purpose is code signing.
+
+Examples of importing this file in your keystore:
+ gpg --import KEYS.txt
+ (need pgp and other examples here)
+
+Examples of adding your key to this file:
+ pgp -kxa  and append it to this file.
+ (pgpk -ll  && pgpk -xa ) >> this file.
+ (gpg --list-sigs 
+ && gpg --armor --export ) >> this file.
+
+---
+pub   4096R/D3541808 2014-01-09
+uid   [ultimate] Suneel Marthi (CODE SIGNING KEY) 
+sig 3D3541808 2014-01-09  Suneel Marthi (CODE SIGNING KEY) 

+sub   4096R/AF46E2DE 2014-01-09
+sig  D3541808 2014-01-09  Suneel Marthi (CODE SIGNING KEY) 

+
+-BEGIN PGP PUBLIC KEY BLOCK-
+Comment: GPGTools - https://gpgtools.org
+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+=f8kJ
+-END PGP PUBLIC KEY BLOCK-
+pub   rsa4096 2017-07-12 [SC]
+  406DCA257CD2BE237B79AE6BC9D353CA4AFF2E24
+uid   [ultimate] Ly Nguyen (CODE SIGNING KEY) 
+sig 3C9D353CA4AFF2E24 2017-07-12  Ly Nguyen (CODE SIGNING KEY) 

+sub   rsa4096 2017-07-12 [E]
+sig  C9D353CA4AFF2E24 2017-07-12  Ly Nguyen (CODE SIGNING KEY) 

+
+-BEGIN PGP PUBLIC KEY BLOCK-
+
+mQINBFlmSIMBEADIr6FzNJ6o/owjqgqWdOtreIRuU47/uzNRZw8c2lEys2Fw+3CI
+iUitkWpb7jR0BGLk+8yUk+1VGdXPuJ+zj8XWcCnCJ7TUy3Hudp/BrX7y388m9hP9
+3LP5yx+AUKbXRZiEr5EG2lyTmJBB5lmreVlRMs74Ie3uFtH6US/DVZMqULEtumcH
+yCL30kKugUjfftO1mbx901kB0WpB705od3Wrde0Jd9sniMz4HkXMsd93gEx

svn commit: r21472 - /dev/incubator/mxnet/0.11.0/ /release/incubator/mxnet/0.11.0/

[nswamy]

Author: nswamy
Date: Tue Sep  5 19:23:17 2017
New Revision: 21472

Log:
0.11.0 release moving from dev to release repo

Added:
release/incubator/mxnet/0.11.0/
  - copied from r21471, dev/incubator/mxnet/0.11.0/
Removed:
dev/incubator/mxnet/0.11.0/

[GitHub] szha commented on a change in pull request #7570: Gluon InstanceNorm and ReflectancePadding

[git]

szha commented on a change in pull request #7570: Gluon InstanceNorm and 
ReflectancePadding
URL: https://github.com/apache/incubator-mxnet/pull/7570#discussion_r137090552
 
 

 ##
 File path: python/mxnet/gluon/nn/basic_layers.py
 ##
 @@ -442,3 +442,87 @@ def hybrid_forward(self, F, x):
 
 def __repr__(self):
 return self.__class__.__name__
+
+
+class InstanceNorm(HybridBlock):
+"""
+Applies instance normalization to the n-dimensional input array.
+This operator takes an n-dimensional input array where (n>2) and 
normalizes 
+the input using the following formula:
+.. math::
+out = \frac{x - mean[data]}{ \sqrt{Var[data]} + \epsilon} * gamma + beta
+This layer is similar to batch normalization layer (`BatchNorm`)
+with two differences: first, the normalization is
+carried out per example (instance), not over a batch. Second, the
+same normalization is applied both at test and train time. This
+operation is also known as `contrast normalization`.
+If the input data is of shape [batch, channel, spacial_dim1, spacial_dim2, 
...],
+`gamma` and `beta` parameters must be vectors of shape [channel].
+This implementation is based on paper:
+.. [1] Instance Normalization: The Missing Ingredient for Fast Stylization,
+   D. Ulyanov, A. Vedaldi, V. Lempitsky, 2016 (arXiv:1607.08022v2).
+Examples::
+  // Input of shape (2,1,2)
 
 Review comment:
   add an empty line after ::
   http://www.sphinx-doc.org/en/stable/rest.html#source-code
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] madjam closed pull request #7678: Clean amalgamation ws and skip failing test

[git]

madjam closed pull request #7678: Clean amalgamation ws and skip failing test
URL: https://github.com/apache/incubator-mxnet/pull/7678
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet] branch master updated: Clean amalgamation ws and skip failing test (#7678)

[madjam]

This is an automated email from the ASF dual-hosted git repository.

madjam pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git


The following commit(s) were added to refs/heads/master by this push:
 new da7c5bc  Clean amalgamation ws and skip failing test (#7678)
da7c5bc is described below

commit da7c5bc7041fddfc411c4e59b2a23204d2532c3a
Author: Sandeep Krishnamurthy 
AuthorDate: Tue Sep 5 12:36:31 2017 -0700

Clean amalgamation ws and skip failing test (#7678)

* Clean amalgamation ws and skip failing test

* Undo skipping the failing test_ce_loss
---
 Jenkinsfile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Jenkinsfile b/Jenkinsfile
index 6adbea1..5b2f819 100644
--- a/Jenkinsfile
+++ b/Jenkinsfile
@@ -164,7 +164,7 @@ try {
   },
   'Amalgamation MIN': {
 node('mxnetlinux') {
-  ws('workspace/amalgamation') {
+  ws('workspace/amalgamationmin') {
 init_git()
 make('cpu', '-C amalgamation/ clean')
 make('cpu', '-C amalgamation/ USE_BLAS=openblas MIN=1')

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" '].

[GitHub] szha commented on a change in pull request #7570: Gluon InstanceNorm and ReflectancePadding

[git]

szha commented on a change in pull request #7570: Gluon InstanceNorm and 
ReflectancePadding
URL: https://github.com/apache/incubator-mxnet/pull/7570#discussion_r137093222
 
 

 ##
 File path: python/mxnet/gluon/nn/basic_layers.py
 ##
 @@ -442,3 +442,90 @@ def hybrid_forward(self, F, x):
 
 def __repr__(self):
 return self.__class__.__name__
+
+
+class InstanceNorm(HybridBlock):
+"""
+Applies instance normalization to the n-dimensional input array.
+This operator takes an n-dimensional input array where (n>2) and 
normalizes 
+the input using the following formula:
+.. math::
+
+out = \frac{x - mean[data]}{ \sqrt{Var[data]} + \epsilon} * gamma + beta
+This layer is similar to batch normalization layer (`BatchNorm`)
 
 Review comment:
   A new line here too, since the block of source code is supposed to be 
surrounded by empty lines.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] zhanghang1989 commented on a change in pull request #7570: Gluon InstanceNorm and ReflectancePadding

[git]

zhanghang1989 commented on a change in pull request #7570: Gluon InstanceNorm 
and ReflectancePadding
URL: https://github.com/apache/incubator-mxnet/pull/7570#discussion_r137095702
 
 

 ##
 File path: python/mxnet/gluon/nn/basic_layers.py
 ##
 @@ -442,3 +442,90 @@ def hybrid_forward(self, F, x):
 
 def __repr__(self):
 return self.__class__.__name__
+
+
+class InstanceNorm(HybridBlock):
+"""
+Applies instance normalization to the n-dimensional input array.
+This operator takes an n-dimensional input array where (n>2) and 
normalizes 
+the input using the following formula:
+.. math::
+
+out = \frac{x - mean[data]}{ \sqrt{Var[data]} + \epsilon} * gamma + beta
+This layer is similar to batch normalization layer (`BatchNorm`)
 
 Review comment:
   Does it look okay now? I am not familiar with rst
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] pracheer opened a new pull request #7746: CoremlBugFixes:Input variable name can be something other than data; ?

[git]

pracheer opened a new pull request #7746: CoremlBugFixes:Input variable name 
can be something other than data; ?
URL: https://github.com/apache/incubator-mxnet/pull/7746
 
 
   ?making pad and stride optional.
   
   Earlier, we were not providing data_names argument while creating the module 
which meant that the input data variable name was assumed to be "data". This is 
fixed. Also, added a unit test for it (due to which utils.load_model(..) had to 
be refactored).
   
   The second bug was we missed assuming pad and stride parameters for 
convolutional layers are optional arguments. Added a unit test for this too.
   
   Also, tested with mnist model from the tutorial page by changing the input 
variable name to something other than "data".
   
   @jiajiechen @srikris @mli @piiswrong 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #7688: Update doc for sparse related APIs

[git]

eric-haibin-lin commented on a change in pull request #7688: Update doc for 
sparse related APIs
URL: https://github.com/apache/incubator-mxnet/pull/7688#discussion_r137112534
 
 

 ##
 File path: python/mxnet/io.py
 ##
 @@ -513,7 +514,8 @@ def _init_data(data, allow_empty, default_name):
 return list(data.items())
 
 class NDArrayIter(DataIter):
-"""Returns an iterator for ``mx.nd.NDArray``, ``numpy.ndarray`` or 
``h5py.Dataset``.
+"""Returns an iterator for ``mx.nd.NDArray``, ``numpy.ndarray``, 
``h5py.Dataset``
+or ``mx.nd.sparse.CSRNDArray``.
 
 Review comment:
   No we don't have any
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] eric-haibin-lin commented on a change in pull request #7688: Update doc for sparse related APIs

[git]

eric-haibin-lin commented on a change in pull request #7688: Update doc for 
sparse related APIs
URL: https://github.com/apache/incubator-mxnet/pull/7688#discussion_r137113966
 
 

 ##
 File path: python/mxnet/io.py
 ##
 @@ -603,6 +617,11 @@ def __init__(self, data, label=None, batch_size=1, 
shuffle=False,
 
 self.data = _init_data(data, allow_empty=False, default_name=data_name)
 self.label = _init_data(label, allow_empty=True, 
default_name=label_name)
+if isinstance(data, CSRNDArray) or isinstance(label, CSRNDArray):
+assert(last_batch_handle == 'discard'), \
+  "`NDArrayIter` only supports ``CSRNDArray`` with `shuffle` 
set to `False`"
 
 Review comment:
   Updated now. 
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] cjolivier01 opened a new pull request #7747: Fix R test problem

[git]

cjolivier01 opened a new pull request #7747: Fix R test problem
URL: https://github.com/apache/incubator-mxnet/pull/7747
 
 
   Fix for: https://github.com/apache/incubator-mxnet/issues/7675
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] cjolivier01 commented on issue #7675: [Jenkins CI] Broken R Unit test during build

[git]

cjolivier01 commented on issue #7675: [Jenkins CI] Broken R Unit test during 
build
URL: 
https://github.com/apache/incubator-mxnet/issues/7675#issuecomment-327304053
 
 
   I think I know what's causing this...
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] kevinthesun opened a new pull request #24: 0.11 updates

[git]

kevinthesun opened a new pull request #24: 0.11 updates
URL: https://github.com/apache/incubator-mxnet-site/pull/24
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] kevinthesun commented on issue #24: 0.11 updates

[git]

kevinthesun commented on issue #24: 0.11 updates
URL: 
https://github.com/apache/incubator-mxnet-site/pull/24#issuecomment-327311173
 
 
   @nswamy Can you merge this plz?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] nswamy closed pull request #24: 0.11 updates

[git]

nswamy closed pull request #24: 0.11 updates
URL: https://github.com/apache/incubator-mxnet-site/pull/24
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[incubator-mxnet-site] branch asf-site updated: 0.11 updates (#24)

[nswamy]

This is an automated email from the ASF dual-hosted git repository.

nswamy pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet-site.git


The following commit(s) were added to refs/heads/asf-site by this push:
 new dbb414a  0.11 updates (#24)
dbb414a is described below

commit dbb414aac7adde466758ef59411fcd903fd2bb6f
Author: Yao Wang 
AuthorDate: Tue Sep 5 15:12:34 2017 -0700

0.11 updates (#24)

* Add README file

* Modify news and gluon link
---
 README.md| 1 +
 index.html   | 4 ++--
 tutorials/index.html | 2 +-
 3 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/README.md b/README.md
new file mode 100644
index 000..e6b7fe6
--- /dev/null
+++ b/README.md
@@ -0,0 +1 @@
+# Apache MXNet Website
diff --git a/index.html b/index.html
index e3b8116..44f4cb6 100644
--- a/index.html
+++ b/index.html
@@ -163,8 +163,8 @@
 http://gluon.mxnet.io";>Learn More
 
 
-MXNet 0.10.0 Released
-We're excited to announce the release of MXNet 0.10.0! Check out the 
release notes for latest updates.
+MXNet 0.11.0 Released
+We're excited to announce the release of MXNet 0.11.0! Check out the 
release notes for latest updates.
 https://github.com/dmlc/mxnet/releases";>Learn More
 
 
diff --git a/tutorials/index.html b/tutorials/index.html
index 8a820b7..35b939d 100644
--- a/tutorials/index.html
+++ b/tutorials/index.html
@@ -182,7 +182,7 @@
 
 Tutorials¶
 These tutorials introduce a few fundamental concepts in deep learning and 
how to implement them in MXNet. The Basics section contains 
tutorials on manipulating arrays, building networks, loading/preprocessing 
data, etc. The Training and Inference section talks about implementing 
Linear Regression, training a Handwritten digit classifier using MLP and CNN, 
running inferences using a pre-trained model, and lastly, efficiently training 
a large scale image class [...]
-Note: We are working on a set of tutorials for the new 
imperative interface called Gluon. A preview version is hosted at http://thestraightdope.mxnet.io";>thestraightdope.mxnet.io.
+Note: We are working on a set of tutorials for the new 
imperative interface called Gluon. A preview version is hosted at http://gluon.mxnet.io";>gluon.mxnet.io.
 
 Python¶
 

-- 
To stop receiving notification emails like this one, please contact
['"comm...@mxnet.apache.org" '].

[GitHub] jiajiechen commented on a change in pull request #7746: CoremlBugFixes:Input variable name can be something other than data; ?

[git]

jiajiechen commented on a change in pull request #7746: CoremlBugFixes:Input 
variable name can be something other than data; ?
URL: https://github.com/apache/incubator-mxnet/pull/7746#discussion_r137129664
 
 

 ##
 File path: tools/coreml/converter/_layers.py
 ##
 @@ -314,7 +314,12 @@ def convert_convolution(net, node, module, builder):
 Wb = None
 
 channels = W.shape[1]
-stride_height, stride_width = literal_eval(param['stride'])
+
+stride_height = 1
+stride_width = 1
+if 'stride' in param.keys():
+stride_height, stride_width = literal_eval(param['stride'])
+
 kernel_height, kernel_width = literal_eval(param['kernel'])
 
 Review comment:
   Shall we do the same initialization for kernel height and kernel width as 
stride height and stride weight?
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] pracheer commented on a change in pull request #7746: CoremlBugFixes:Input variable name can be something other than data; ?

[git]

pracheer commented on a change in pull request #7746: CoremlBugFixes:Input 
variable name can be something other than data; ?
URL: https://github.com/apache/incubator-mxnet/pull/7746#discussion_r137131017
 
 

 ##
 File path: tools/coreml/converter/_layers.py
 ##
 @@ -314,7 +314,12 @@ def convert_convolution(net, node, module, builder):
 Wb = None
 
 channels = W.shape[1]
-stride_height, stride_width = literal_eval(param['stride'])
+
+stride_height = 1
+stride_width = 1
+if 'stride' in param.keys():
+stride_height, stride_width = literal_eval(param['stride'])
+
 kernel_height, kernel_width = literal_eval(param['kernel'])
 
 Review comment:
   Thanks for reviewing Jiajie.
   
   Kernel is a required parameter for the de/convolutional layer so we can be 
certain that it'll exist whereas the same thing can't be said about pad/stride. 
See more 
**[here](https://mxnet.incubator.apache.org/api/python/symbol.html#mxnet.symbol.Convolution).**
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mbaijal opened a new issue #7748: MXNet 0.11.0 Release Feedback: NOTICE file

[git]

mbaijal opened a new issue #7748: MXNet 0.11.0 Release Feedback: NOTICE file
URL: https://github.com/apache/incubator-mxnet/issues/7748
 
 
   All added Notices should be removed from the top level NOTICE file as 
indicated in this comment: 
   "Unless the dependencies listed have explicit notice requirements (they 
don't seem to) nothing needs to be added to the NOTICE file"
   
   Discussed in this email thread: 
https://lists.apache.org/thread.html/171c0ff73e5f7539611a8024fef42968f5ffea800c354b2cdfa04ab2@%3Cgeneral.incubator.apache.org%3E
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mbaijal opened a new issue #7749: MXNet 0.11.0 Release Feedback: Source Headers and License Files

[git]

mbaijal opened a new issue #7749: MXNet 0.11.0 Release Feedback: Source Headers 
and License Files
URL: https://github.com/apache/incubator-mxnet/issues/7749
 
 
   **1. Headers**
   - "The source headers should be reverted in any areas where not all 
contributors to that file have signed ICLAs, otherwise we need some kind of 
agreement they are OK with moving to NOTICE file (which is how this has been 
handled in the past)"
   - While a number of file (around 50 odd) do have "Copyright (c) 2016 by 
Contributors? they don?t have a full Apache licensed header please fix this in 
the next release. More files exist with other years and without the correct 
header.
   
   **2. License Files**
   - Missing - BSD license for this file [4]
   - Missing - MIT licensed JQuery and Sizzle.js [5]
   - Missing - This BSD licensed file [6]
   - Missing - Boost Software licensed file [7]
   - Missing - This BSD licensed files [8][9]
   - This license file may be a little misleading [1] where its apache license 
but also claims ?All rights reserved?. I notice this in a number of other 
places as well.
   - A number of Apache licenses have incorrect appendixes with the copyright 
line filled in e.g. [2]
   - This file may incorrect have an ASF header on it [3]
   
   **3. Files referred above can be found as follows** - 
   [1] ./src/operator/contrib/ctc_include/LICENSE
   [2]../dlpack/LICENSE
   [3] 
./example/image-classification/predict-cpp/image-classification-predict.cc
   [4] ./docs/_static/searchtools_custom.js
   [5] ./docs/_static/jquery-1.11.1.js
   [6] ./cub/test/mersenne.h
   [7] ./cmake/Modules/FindJeMalloc.cmake
   [8] ./dmlc-core/cmake/Modules/FindCrypto.cmake
   [9] ./nnvm/dmlc-core/cmake/Modules/FindCrypto.cmake
   
   **4. Links to the email threads which discuss these issues -** 
   1. 
https://lists.apache.org/thread.html/78b01bc01e107300257ca2bc7ad04c7eae8c87bcc4cffaaf829813bd@%3Cgeneral.incubator.apache.org%3E
   2. 
https://lists.apache.org/thread.html/394916505f0f1e157dbb650d61caed2e4518ff3dc49e20a4db239779@%3Cdev.mxnet.apache.org%3E
   3. 
https://lists.apache.org/thread.html/d88544cc13275c502a1580b16b0575f9d827a0041c0bd2879b7a9197@%3Cdev.mxnet.apache.org%3E
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mbaijal opened a new issue #7750: MXNet 0.11.0 Release Feedback: README File

[git]

mbaijal opened a new issue #7750: MXNet 0.11.0 Release Feedback: README File
URL: https://github.com/apache/incubator-mxnet/issues/7750
 
 
   Copied and assembled as is from this [email thread] 
(https://lists.apache.org/thread.html/171c0ff73e5f7539611a8024fef42968f5ffea800c354b2cdfa04ab2@%3Cgeneral.incubator.apache.org%3E)
   
   1. Install Instructions needed - "Agreed with Justin that there needs to be 
a Getting Started text file of some kind. This could be a link from the 
README.md to the docs/get_started directory perhaps; though .md format isn't 
the easiest to read (ie: link to website and over time we should consider 
whether a local version is needed)."
   2. Also the README has a large number of links to http://mxnet.io. These 
should be changed.
   3. R-package/ directory is empty (needs removing as confusing, or a README 
could be added explaining the code is not present and can be found outside of 
Apache).
   4. The README.md also refers to rc3. It shouldn't refer to rc3 as a release, 
and ideally it would refer to 0.11.0 as a release (though tricky to be forward 
looking given that GitHub treats that as a homepage). Randomly noting that the 
What's New should be dated.
   5. The README.md refers to the copyright being owned by Contributors. Needs 
updating to a license statement (with NOTICE handling the copyright side of 
things).
   
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] mbaijal opened a new issue #7751: MXNet 0.11.0 Release Feedback: NEWS.md and CONTRIBUTORS.md

[git]

mbaijal opened a new issue #7751: MXNet 0.11.0 Release Feedback: NEWS.md and 
CONTRIBUTORS.md
URL: https://github.com/apache/incubator-mxnet/issues/7751
 
 
   As pointed out in this [email 
thread](https://lists.apache.org/thread.html/171c0ff73e5f7539611a8024fef42968f5ffea800c354b2cdfa04ab2@%3Cgeneral.incubator.apache.org%3E)
   1. CONTRIBUTORS.md calls the project DMLC/MXNet.
   2. The NEWS.md refers to 0.11.0-rc3 as the latest version. This should refer 
to the version being released rather than the rc3.
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji opened a new pull request #7752: Use memcpy instead of assigning each individual element

[git]

yaochengji opened a new pull request #7752: Use memcpy instead of assigning 
each individual element
URL: https://github.com/apache/incubator-mxnet/pull/7752
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji closed pull request #7752: Use memcpy instead of assigning each individual element

[git]

yaochengji closed pull request #7752: Use memcpy instead of assigning each 
individual element
URL: https://github.com/apache/incubator-mxnet/pull/7752
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji opened a new pull request #7753: Use memcopy of assigning each individual element

[git]

yaochengji opened a new pull request #7753: Use memcopy of assigning each 
individual element
URL: https://github.com/apache/incubator-mxnet/pull/7753
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

[GitHub] yaochengji closed pull request #7753: Use memcopy of assigning each individual element

[git]

yaochengji closed pull request #7753: Use memcopy of assigning each individual 
element
URL: https://github.com/apache/incubator-mxnet/pull/7753
 
 
   
 

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services